• Yu-Fei Zhan, Feng Li, Long-Chuan Wu, Lin Chen, Can-Yan Zhu, Ming-Shuai Han, Guo-Fang Ma, and Yong-Hong Zhong.
• Emergency Medicine, Linping Campus, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, Zhejiang, China.
• Medicine (Baltimore). 2024 Nov 1; 103 (44): e40033e40033.

AbstractThe purpose of this study was to investigate the value of the Charlson comorbidity index (CCI) in predicting intensive care unit (ICU) readmission in aortic aneurysm (AA) patients. Patient information came from the Medical Information Mart for Intensive Care- IV (MIMIC-IV) database. The relationship between CCI and ICU readmission was analyzed by restricted cubic spline, generalized linear regression, trend analysis, and hierarchical analysis. The clinical value of CCI in predicting ICU readmission was analyzed by receiver operating characteristic curve, decision curve analysis, XGBoost regression, and random forest regression. A total of 523 patients with AA were enrolled in the study. Patients with AA who were readmitted to the ICU had higher width of red blood cell distribution width (RDW) and higher CCI. CCI had better performance and clinical net benefit for predicting ICU readmission than RDW. An independent nonlinear relationship was found between CCI and ICU readmission. The trend analysis suggested that the risk of ICU readmission increased with higher CCI scores. The hierarchical analysis showed that their association was mainly found in surgery requirement populations regardless of AA types. Further, CCI was found to have better clinical value in predicting ICU readmission of thoracic aortic aneurysm (TAA) patients undergoing surgery. Age, renal disease, chronic lung disease, and dementia were important components of CCI in predicting ICU readmission of TAA patients undergoing surgery. CCI was independently associated with the ICU readmission of AA patients in a positive relationship and had more favorable prediction performance in TAA patients who underwent surgery.Copyright © 2024 the Author(s). Published by Wolters Kluwer Health, Inc.

18 keywords...

hide…